One of the most important problems in psychiatry today is that of differentiating bipolar II from unipolar depression. Misdiagnosing an episode of bipolar depression results in the initiation of inappropriate treatment, which at best may be ineffective and at worst may exacerbate the course of illness. Thus, biomarkers that could accurately differentiate unipolar from bipolar II depression would greatly enhance our ability to provide outstanding mental health treatment for veterans. Functional neuroimaging methodologies are maturing rapidly and this technology may eventually provide reliable clincally useful biomarkes that can differentiate mood disorders. In order to achieve that goal, we need a better understanding of the neurobiological differences and similiarites between bipolar II and unipolar depression. Further, we need to develop functional neuroimaging activation pardigms that have the potential to be clincially useful. The proposed study will test the following hypotheses which will address both of those goals: (1) Bipolar II depression is neurobiologically distinct from unipolar depression. Further, differences between the disorders are manifested, at least in part, in striatal function as a result of differential dopamine modulation. Therefore, activation of the striatum and functional connectivity of this region with cortical and subcortical areas will be significantly different in bipolar II versus unipolar depression. Specifically in bipolar II depression, there will be increased activation of the striatum and increased connectivity of the striatum with other regions while in unipolar depression there will be decreased activation and connectivity. These differences will be manifested in both comparisons of bipolar and unipolar depressed subjects to controls as well as comparisons of the two illness groups to each other. (2) Functional magnetic resonance imaging (fMRI) using motor activation paradigms will demonstrate robust differences in striatal activation and connectivity between bipolar II and unipolar depression. This methodology may be refined in future studies to the point that it will be clinically useful as a biomarker to differentiate unipolar from bipolar II depression. In order to evaluate these hypotheses we propose an fMRI study with the following aims: (1) Identify patterns of striatal motor activation and task-related connectivity of both bipolar II and unipolar depression as compared to controls. (2) Determine differences and similarities of patterns of striatal motor task-related activation and connectivity for bipolar II depression versus unipolar depression. (3) Determine which of the three motor tasks provide the most robust differentiation between bipolar II and unipolar depression based on statistical comparisons of volumes of activation for the striatum and corticostriatal circuitry. The proposed project will be a single site, cross-sectional functional neuroimaging study comparing 30 males with unipolar depression;30 males with bipolar II depression and 30 matched male controls without neuropsychiatric illness. PUBLIC HEALTH RELEVANCE: A recent retrospective analysis reported that 65,556 veterans have bipolar disorder, which represents 33.3 % of the individuals with serious mental illness who receive services from the Department of Veterans Affairs. These data provide compelling evidence that the treatment of bipolar disorder represents a tremendous institutional challenge for the VA. Improving our ability to accurately diagnose this condition can greatly enhance our ability to provide appropriate treatment for veterans with bipolar illness.